Journal of Surgical Research 138, 175–180 (2007)

Association for Academic Surgery, 2006

Angiotensin II Type 2 Receptor Decreases Ischemia ReperfusionInduced Fluid Leak1

René Ramirez, M.D., Terry Chong, M.D., Brian Curran, B.S., Javid Sadjadi, M.D.,and Gregory P. Victorino, M.D.2

Department of Surgery, UCSF-East Bay, Alameda County Medical Center, Alameda, California

Submitted for publication January 9, 2006

doi:10.1016/j.jss.2006.07.049

Background. Intravascular volume loss from ische-mia-reperfusion (IR) injury is a major clinical con-cern. We hypothesize that angiotensin II decreasesIR-mediated microvascular fluid leak via the angio-tensin II type 2 receptor in a cAMP dependent manner.We therefore sought to determine hydraulic perme-ability after IR of venules treated with 1) angiotensinII, 2) angiotensin II and cAMP synthesis inhibitor, and3) angiotensin II and an angiotensin II type 2 receptorantagonist.

Methods. Rat mesenteric post-capillary venules weremicro-cannulated to measure hydraulic permeability(Lp). IR was achieved by placing animals in a 5% oxy-gen environment and preventing venular flow, afterwhich blood flow was allowed to resume. Lp was mea-sured after IR and treatment with 1) angiotensin II(20 nM), 2) angiotensin II (20 nM) � cAMP synthesisinhibitor (DDA,10uM), and 3) angiotensin II (20 nM) �type 2 receptor antagonist (PD-123319, 300 �M) (n � 6in each group).

Results. Compared with the seven-fold increase inLp because of IR alone: 1) angiotensin II attenuatedthe seven-fold increase by 50% (P < 0.005), 2) cAMPinhibition did not change the effect of angiotensin IIon leak, and the type 2 receptor antagonist completelyblocked the effects of angiotensin II on IR mediatedleak.

Conclusion. Treatment with angiotensin II attenu-ated increases in hydraulic permeability because of IRby 50%. Inhibition of cAMP synthesis did not changethe effect of angiotensin II on IR mediated leak, andangiotensin II type 2 receptor inhibition completelyblocked the effects of angiotensin II. These resultsindicate that angiotensin II decreases IR induced leak

1 Presented at the 1st Annual Academic Surgical Congress (Asso-ciation for Academic Surgery), San Diego, CA, February 7–11, 2006.

2 To whom correspondence and reprint requests should be ad-dressed at Department of Surgery, UCSF-East Bay, 1411 East 31st

street, Oakland, CA 94602. E-mail: victorinog@surgery.ucsf.edu.

175

via the angiotensin II type 2 receptor in a cAMP inde-pendent manner. A better understanding of mediatorsthat reduce intravascular fluid loss from IR-inducedmicrovascular dysfunction may help clinicians treatuncontrolled fluid extravasation that occurs duringshock and sepsis. © 2007 Elsevier Inc. All rights reserved.

Key Words: ischemia reperfusion; angiotensin II; AT2receptor; cAMP; microvascular permeability.

INTRODUCTION

Ischemia/reperfusion (IR) affects many physiologicalvariables, including endothelial barrier function. Thiscompromise of endothelial-barrier integrity leads to theloss of intravascular fluid into the interstitium and isa major cause of increased morbidity and mortality incritically ill patients [1, 2]. The majority of IR-inducedendothelial barrier dysfunction occurs at the post-cap-illary venule because of changes in endothelial cell func-tion [3, 4].

Derangements in endothelial cell function includealterations in the release of vasoactive mediators andsecond messenger cyclic nucleotide homeostasis. An-giotensin II levels are increased during times of phys-iological stress [5]. Previous investigations found thatresting post-capillary venules exposed to angiotensin IIled to a five-fold increase in hydraulic permeability [6].Conversely, when a vessel is activated with a substancesuch as ATP, subsequent angiotensin II exposure hasthe opposite effect, which is to decrease microvascularpermeability [6]. The dual effect of angiotensin II onmicrovascular permeability may be the result of dif-ferential activation of the two subtype receptors (an-giotensin type 1 and angiotensin type 2), which mayvary depending on the state of the post-capillary venule.The angiotensin II subtype receptors have differenteffects on second messenger levels, which could ac-count for the opposing actions of the two receptors andthe dual effects of angiotensin II on microvascular per-

meability [7, 8].

0022-4804/07 $32.00© 2007 Elsevier Inc. All rights reserved.

176 JOURNAL OF SURGICAL RESEARCH: VOL. 138, NO. 2, APRIL 2007

Previous studies examining the role of second mes-senger cyclic nucleotides in microvascular permeabil-ity homeostasis, demonstrate that under basal/restingconditions, cAMP decreases fluid leak [9]. Consider-ing the dual effect of angiotensin II on microvascularpermeability and evidence that angiotensin II affectscAMP, we chose to investigate the effect of angioten-sin II on IR-mediated leak and examine its interactionwith cAMP. We hypothesize that angiotensin II de-creases IR-mediated microvascular fluid leak via theangiotensin II type 2 receptor in a cAMP dependentmanner. The purposes of our study are to: 1) examinethe effect of angiotensin II on IR-induced increases inmicrovascular permeability; 2) explore the interactionof angiotensin II with cAMP during post-capillaryvenular IR; and 3) examine the role of the angiotensinII type-2 receptor in IR-induced increases in microvas-cular permeability.

MATERIALS AND METHODS

All studies were approved and complied with institutional animalresearch protocols. Preparation of the animals and the Ringer’ssolution has been described previously [6]. They are briefly describedbelow.

Solution Preparation

RBCs used as flow markers were harvested from female GoldenSyrian hamsters (140–180 g; Harlan, Indianapolis, IN.). The bloodwas centrifuged to remove the plasma and buffy coat; then washedthree times in 15 mL of mammalian Ringer’s solution.

The mammalian Ringer’s solution was prepared daily in dis-tilled deionized water and contained 135 mM NaCl; 4.6 mM KCl;2.0 mM CaCl; 2.46 mM MgS04; 5.0 mM NaHCO3; 5.5 mM Dextrose;9.03 mM HEPES Salt (Research Organics, Cleveland, OH); and 11.04mM HEPES Acid (Research Organics). Bovine serum albumin (BSA)was added to the Ringer’s to prepare a 1% BSA Ringer’s solution forperfusion (BSA crystallized; Sigma Chemical, St. Louis, MO).

The test perfusates consisted of hamster RBC markers, 1% BSAsolution, and test mediator(s). The mediators included angiotensin II(20 nM) (Sigma), a selective angiotensin II type 2 receptor antagonist,PD-123319, (300 �M) (Sigma), and 10 �M 2,5=-dideoxyadenosine(ddA, a highly specific adenylate cyclase inhibitor that prohibitscAMP synthesis) (Biomol, Plymouth Meeting, PA). The doses forangiotensin II, PD-123319, and 2=,5=ddA, were based on previouslypublished doses [6, 9, 10].

Animal Preparation

Adult female Sprague-Dawley rats (250–310 g; Hilltop Lab Ani-mals Inc., Scottsdale, PA) were anesthetized with subcutaneous so-dium pentobarbital (60 mg/kg body weight). The bowel and mesen-tery was exposed and positioned on an inverted microscope stage(Diaphot; Nikon, Melville, NY). The animal’s body temperature wasmaintained at 37°C throughout the study. The mesentery was con-tinuously bathed in mammalian Ringer’s solution.

Rat mesenteric post-capillary venules, 20 to 30 �m in diameterand at least 400 �m in length, were identified based on flow patterns.Vessels with no evidence of leukocyte adherence or side brancheswere chosen. The vessels were cannulated with micropipettes at-tached to a water manometer for control of hydrostatic perfusion

pressure.

Measurement of Hydraulic Permeability

Single vessel Lp was determined using the modified Landis micro-occlusion technique. The assumptions and limitations of this modelhave been previously described [11]. Initial cell velocity (dl/dt) wasdetermined by recording marker cell position as a function of time.Transmural water flux per unit area (Jv/S) was calculated by theequation: Jv/S � (dl/dt)(r/21), where r is the capillary radius and 1 isthe initial distance between the marker cell and occluded site. De-termination of hydraulic permeability (Lp) was based on a modifiedversion of Starling’s equation of fluid filtration: Lp � (Jv/S)(1/Pc),where Pc is the capillary hydrostatic pressure. Control studies havedocumented the stability of this model over time and after multiplecannulations of the vessels [6, 11].

EXPERIMENTAL DESIGN

IR Protocol

Ischemia was produced by preventing blood flowthrough the study microvessel in a 10% oxygen envi-ronment for 15 min. First, the rat was placed in asealed container infused with nitrogen gas to lowerfractional inspired oxygen to 10%. Next, the superfu-sate bathing the mesentery was bubbled with nitrogengas to lower oxygen to 10%. Last, perfusate solutions inthe micropipettes were made hypoxic by bubbling ni-trogen gas into the solution before microvessel cannu-lation. Oxygen probes were used to ensure 10% atmo-spheric oxygen. Oxygen probes were placed inside thecontainer, directly on the mesentery, and next to thestudy vessel to measure fractional oxygen content, andto ensure maintenance of the 10% oxygen environ-ment.

Reperfusion was achieved by returning normal bloodflow through the microvessel for 35 min while simul-taneously achieving reoxygenation in the atmosphere,superfusate, and perfusate. Again, probes were used todocument oxygen content of 21%.

Measurements of Lp were determined during base-line, after ischemia, and again after 15 min of reperfu-sion. Subsequent measurements were also made afterreperfusion at 5-min intervals over a final 20-min pe-riod.

IR and Angiotensin II

To investigate the effect of angiotensin II on IR-induced increases in post-capillary venular fluid leak,the IR protocol was followed as outlined above. How-ever, during the reperfusion period, 20 nM of angioten-sin II was perfused into the study vessel. Measure-ments of Lp were obtained at the end of the ischemiaperiod and at 5-min intervals during the 35-min reper-fusion period (n � 6).

IR and Angiotensin II During cAMP Inhibition

To explore whether exogenous angiotensin II treat-ment acts via the stimulation of cAMP during post-

capillary venular IR, cAMP synthesis was inhibited

177RAMIREZ ET AL.: ANGIOTENSIN II AND IR FLUID LEAK

by injecting the adenylate cyclase inhibitor, 2,5=-dideoxyadenosine (10 �M), into the study microvesselduring the ischemia period. This would prevent angio-tensin II-stimulated cAMP synthesis during angioten-sin II treatment. The reperfusion period followed, dur-ing which 20 nM of angiotensin II was perfused into thestudy vessel. Measurements of Lp were obtained at theend of the ischemia period and at 5-min intervals dur-ing the 35-min reperfusion period (n � 6).

IR and Angiotensin II With Angiotensin Type 2Receptor Antagonist

To explore whether angiotensin II treatment acts viathe activation of the angiotensin type 2 receptor duringpost-capillary venular IR, the angiotensin type 2 recep-tor was blocked by injecting the receptor antagonist,PD-123319 (300 �M), into the study microvessel duringthe ischemia period. This prevents binding of angioten-sin II to the type 2 receptor after the administration ofangiotensin II. The reperfusion period followed, duringwhich 20 nM of angiotensin II was perfused into thestudy vessel. Measurements of Lp were obtained at theend of the ischemia period and at 5-min intervals dur-ing the 35-min reperfusion period (n � 6).

Statistical Analysis

Group means of sequential measurements were an-alyzed by repeated measures ANOVA with post-hocanalysis. Measurements between different study groupswere analyzed using unpaired Student’s t-test andANOVA with Bonferroni post-hoc analysis. Statisticalsignificance was set at an alpha error of 5%. All values

FIG. 1. The effect of IR on hydraulic permeability (Lp). Ischemiaelevated Lp approximately 2.5-fold from baseline and reperfusionsubsequently increased Lp by over seven-fold. *Significant increaseversus baseline (P � 0.01). Error bars denote � SEM; L units are

p

�10�7cm/sec/cmH2O.

for Lp are represented as mean � SEM � 10�7 cm/s�1/cmH2O

�1 (n � 6 for all groups).

RESULTS

Effect of IR on Lp

Ischemia for 15 min elevated Lp approximately 2.5-fold from baseline (P � 0.005). Reperfusion increasedLp by over seven-fold with a peak at 15 min of reper-fusion (Lp � 7.07 � 0.20) (P � 0.001). After 15 min ofreperfusion, Lp began to steadily decrease, but after 35min of reperfusion, it still remained elevated at 1.81 �0.10 compared with baseline (P � 0.01) (Fig. 1).

Effect of IR and Angiotensin II on Lp

Angiotensin II treatment during reperfusion signifi-cantly attenuated Lp by 52% (3.36 � 0.18) at 15 min ofreperfusion compared with IR alone (Lp � 7.07 � 0.20)(P � 0.005). After 15 min, the Lp began to decrease ata rate similar to that of IR alone (Fig. 2).

Effect of IR and Angiotensin II During cAMP Inhibition

The cAMP inhibitor did not change the effect of an-giotensin II treatment. After 15 min of reperfusion, theLp for angiotensin II alone was not significantly dif-ferent from the Lp for angiotensin II during cAMPinhibition (3.36 � 0.18 versus 3.37 � 0.18; P � 0.221)(Fig. 3). Cyclic AMP inhibition alone decreased Lp to4.47 � 0.2 from IR alone (Lp � 7.07 � 0.20) at 30-min

FIG. 2. The effect of angiotensin II on IR induced increases onhydraulic permeability (Lp). Angiotensin II treatment attenuated IRmediated increases in Lp by 52% (3.36 � 0.18) at 15 min of reperfu-sion compared with IR alone (Lp � 7.07 � 0.20) (P � 0.005). *Sig-nificant decrease versus IR (P � 0.005). Error bars denote � SEM;Lp units are �10�7cm/sec/cmH2O.

of reperfusion (P � 0.01).

178 JOURNAL OF SURGICAL RESEARCH: VOL. 138, NO. 2, APRIL 2007

Effect of IR and Angiotensin II With Angiotensin Type 2Receptor Antagonist

The angiotensin type 2 receptor antagonist com-pletely blocked any decrease in Lp because of angioten-sin II treatment. During IR alone, Lp peaked at 7.07 �0.20 after 15-min of reperfusion. During angiotensin IItreatment plus type 2 receptor antagonism, Lp peakedat 6.44 � 0.65 (P � 0.155). After 15-min, Lp began todecrease at a rate similar to that for IR alone (Fig. 4).

FIG. 4. The effect of angiotensin II type 2 receptor antagonist onIR induced increases on hydraulic permeability (Lp). The angiotensinII type 2 receptor antagonist completely blocked any decrease angio-tensin II had on IR induced leak. There was no difference at all timepoints between the IR alone group and the angiotensin II plus theangiotensin II type 2 receptor antagonist during IR group (P � 0.15).This indicates that angiotensin II decreases IR induced fluid leak viathe angiotensin II type 2 receptor. Error bars denote � SEM; L units

FIG. 3. The effect of angiotensin II and cAMP inhibition on IRinduced increases on hydraulic permeability (Lp). The effect of thecAMP inhibitor plus angiotensin II did not differ at all time pointsfrom that of angiotensin II alone (P � 0.22). This indicates thatangiotensin II does not decrease IR induced fluid leak via cAMP.Error bars denote � SEM; Lp units are �10�7cm/sec/cmH2O.

p

are �10�7cm/sec/cmH2O.

For comparison, the IR data, angiotensin II treat-ment during IR data, and angiotensin II plus angioten-sin II type 2 receptor blockade during IR data aredisplayed together in Fig. 5. For each group, the datawere completed as separate experiments.

DISCUSSION

Post-capillary venular endothelial cells are espe-cially susceptible to the effects of IR with immediatefunctional alterations [12]. We attempted to furtherdefine the functional changes that occur in the endo-thelial cell that are associated with IR mediated fluidleak. The majority of IR studies investigating endothe-lial cell changes have been in cell culture, which doesnot allow for the influences of important factors suchas paracrine functions of local auxiliary cells. Also,in vitro studies reveal that the permeability of endo-thelial cell monolayers to macromolecules is far greaterthan that measured in vivo [13]. Therefore, we chosean in vivo model that directly measures fluid leak inpost-capillary venules to examine the role of angioten-sin II, the interaction of angiotensin II with cAMP, andthe role of the angiotensin II type-2 (AT2) receptor inIR induced increases in microvascular permeability.

FIG. 5. Angiotensin II treatment of IR induced increases onhydraulic permeability (Lp). Angiotensin II treatment attenuated IRmediated increases in Lp by 52% at 15 min of reperfusion comparedwith IR alone (P � 0.005). The angiotensin II type 2 receptor antag-onist completely blocked any decrease angiotensin II had on IRinduced leak (P � 0.15). These data suggest that angiotensin IIdecreases IR induced fluid leak via the angiotensin II type 2 receptor.The data for each group was completed as separate experiments andare shown here together for comparison. *Significant difference ver-sus angiotensin II and the angiotensin II type 2 receptor antagonist(P � 0.005). Error bars denote � SEM; Lp units are �10�7cm/sec/cmH2O.

The findings of this study indicate that angiotensin II

179RAMIREZ ET AL.: ANGIOTENSIN II AND IR FLUID LEAK

decreases IR-mediated increases in fluid leak via theAT2 receptor, independent of cAMP.

Deciphering the role of angiotensin II during IR-me-diated changes in endothelial barrier function is par-ticularly important considering that angiotensin II issynthesized and released by endothelial cells acting inan autocrine/paracrine manner during times of physi-ological stress [14, 15]. On the endothelial cell surface,angiotensin II activates two unique, G- protein associ-ated, seven transmembrane receptors, the type-1 (AT1)and type-2 (AT2) receptor subtypes [16]. Separate stud-ies have demonstrated that AT1 receptor activation re-sults in an increase in permeability, whereas AT2 re-ceptor activation leads to a decrease in microvascularfluid leak [10, 17]. AT1 receptor stimulation leads to theactivation of protein kinase C (PKC) and an increase inintracellular calcium ([Ca2�]i) [18]. The AT1 receptoralso activates the mitogen-activated protein (MAP) ki-nase cascade and increases intracellular levels cGMP[7, 19]. Each of these mechanisms—increased [Ca2�]i,activation of the MAP kinase cascade, and increasedintracellular cGMP—are related to increased endothe-lial hydraulic permeability [20]. These second messen-ger systems are thought to exert their effect on micro-vascular permeability by phosphorylating myosin lightchain kinase (MLCK), which, in turn, results in con-traction of the cytoskeleton and widening of intercellu-lar gap junctions [20].

In contrast, activation of the AT2 receptor has beenlinked to decreases in calcium influx and cGMP levels,[8] both of which are associated with decreased micro-vascular permeability [20]. Furthermore, AT2 stimula-tion is associated with activation of phosphotyrosinephosphatases [21], which could result in dephosphory-lation of MLCK, thereby decreasing endothelial perme-ability. The evidence points to opposing effects of theangiotensin II receptor subtypes in which activation ofAT1 increases microvascular permeability and AT2 de-creases it. Our finding that the selective activation ofthe AT2 receptor led to significant attenuation of IR-induce fluid leak by angiotensin II further supports theopposing actions of the angiotensin II subtype recep-tors.

The selective activation of the AT2 receptor by an-giotensin II during IR could be explained by a negativefeed-back mechanism. One of the changes that occur atthe endothelial cell level during IR is an increase inintracellular levels of calcium [22]. We have previouslydemonstrated that the AT1 receptor mediates the in-creases in microvascular permeability associated withangiotensin II [17]. Furthermore, we found that theAT1 receptor increases permeability in a calcium de-pendent process [23]. The ischemic component of ourstudy, followed by reperfusion, would increase endo-thelial cell calcium levels, which in turn could exert a

negative feed back mechanism on the AT1 receptor

resulting in the selective activation of the AT2 receptorand the resultant decrease in IR-induced leak seen inour study. This may occur through calcium-mediatedactivation of phospholipase C, resulting in the activa-tion of protein kinase C [24]. Protein kinase C thenphosphorylates the AT-1 receptor and uncouples theAT-1 receptor from G proteins, leading to the deacti-vation of the AT-1 receptor [25].

To further characterize the endothelial cell changesthat occur in post-capillary venules during IR-medi-ated changes in microvascular permeability, we exam-ined the interaction of angiotensin II with cAMP. Cy-clic AMP is associated with decreases in microvascularpermeability, and the AT1 and AT2 receptors decreaseor increase cAMP levels, respectively [7, 8]. We hypoth-esized that angiotensin II decreased IR-induced fluidleak by increasing cAMP levels. Administration of thecAMP inhibitor during ischemia would be expected tonegate the permeability decreasing effects of angioten-sin II treatment during reperfusion. However, we foundthat it had no effect. This result indicates that angio-tensin II decreases IR-mediated increases in microvas-cular permeability, independently of cAMP.

The actions of angiotensin II during IR in decreasingmicrovascular leak could be explained by alterationsin permeability factors other then cAMP. We demon-strated that angiotensin II decreases leak via the AT2receptor. Activation of the AT2 receptor is associatedwith decreased cGMP levels, decreased calcium lev-els, and activation of phosphotyrosine phosphatases,all of which are associated with decreased microvascu-lar permeability [8, 20, 21]. Cyclic GMP can modulatecAMP activity. Cyclic GMP activates a cGMP-depen-dent cAMP phosphodiesterase, which lowers cAMPlevels and thereby increases microvascular permeabil-ity [26]. If angiotensin II was exerting its permeability-decreasing effects by decreasing cGMP and therebydecreasing cAMP phosphodiesterase activity and in-creasing cAMP levels, the resultant decrease in IR-mediated leak via angiotensin II would be negated bythe cAMP synthesis inhibitor. However, the cAMP syn-thesis inhibitor had no effect on the capacity of angio-tensin II to decrease IR-mediated leak. A possible ex-planation may be in the activation of phosphotyrosinephosphatases. One of the results of AT2 receptor bind-ing is phosphotyrosine phosphatase activation, whichcould lead to the dephosphorylation of MLCK, which isassociated with a decrease in endothelial permeability[20, 21]. This activation occurs downstream of cAMP,which could explain why angiotensin II decreases IR-induced leak independent of cAMP.

IR continues to contribute to the morbidity and mor-tality of the critically ill patient [27]. Many importantpathophysiologic changes occur at the post-capillaryvenule. More specifically, these changes can be attrib-

uted to alterations at the level of the post-capillary

180 JOURNAL OF SURGICAL RESEARCH: VOL. 138, NO. 2, APRIL 2007

venular endothelial cell. We have demonstrated thatIR induces a rapid increase in post-capillary venularfluid leak that is significantly attenuated by angioten-sin II. During IR, the permeability-decreasing effect ofangiotensin II is mediated by the angiotensin type 2receptor by what appears to be a process that is inde-pendent from cAMP. Angiotensin II may exert finecontrol over endothelial cell function by selective acti-vation of the type 1 versus type 2 angiotensin receptor.This selective activation is dependent on the stateof the post-capillary venule, with the type 1 receptoractivated in a resting or basal state and the type 2receptor activated during IR.

REFERENCES1. Carden DL, Young JA, Granger DN. Pulmonary microvascular

injury after intestinal ischemia-reperfusion: Role of P-selectin.J Appl Physiol 1993;75:2529.

2. Xiao F, Eppihimer MJ, Young JA, et al. Lung Neutrophil re-tention and injury following intestinal ischemia-reperfusion.Microcirculation 1997;4:359.

3. Granger DN. Ischemia-reperfusion: Mechanisms of microvascu-lar dysfunction and the influence of risk factors for cardiovas-cular disease. Microcirculation 1999;6:167.

4. Granger DN. Physiology and pathophysiology of the microcir-culation. Prog Cardiovasc Med 1998;3:123.

5. Hilton JG, Marullo DS. Trauma induced increases in plasmavasopressin and angiotensin II. Life Sc 1987;41:2195.

6. Victorino GP, Newton CR, Curran B. Effect of angiotensin II onmicrovascular permeability. J Surg Res 2002;104:77.

7. Caputo L, Benessiano J, Boulanger CM, Levy BI. Angiotensin IIincreases cGMP content via endothelial angiotensin II ATI sub-type receptors in the rat carotid artery. Arterioscler ThrombVasc Biol 1995;15:1646.

8. De Gasparo M. AT(1) and AT(2) angiotensin II receptors: Keyfeatures. Drugs 2002;62:Spec No 1:1.

9. Chong TJ, Victorino GP. Cyclic nucleotide second messengers(cAMP and cGMP) play a central role in signal transduction andregulation of mesenteric postcapillary fluid leak. J Trauma2005;59:302.

10. Newton CR, Curran B, Victorino GP. Angiotensin II type 2receptor effect on microvascular hydraulic permeability. J SurgRes 2004;120:83.

11. Curry FE, Huxley VH, Sarelius IH. Measurement of permeabil-ity, pressure, and flow. In: Linden RT, eds. Cardiovascularphysiology. Techniques in life sciences, Vol. P 309. New York:

Elsevier; 1983:1–34.

12. Carden D, Granger DN. Pathophysiology of ischaemia-reper-fusion injury. J Pathol 2000;190:255.

13. Kvietys PR, Granger DN. Endothelial cell monolayers as a toolfor studying microvascular pathophysiology. Am J Physiol 1997;273:G1189.

14. Kusaka Y, Kelly RA, Williams GH, Kifor AI. Coronary micro-vascular endothelial cells cosecrete angiotensin II and endo-thlein-1 via a regulated pathway. Am J Physiol Heart CircPhysiol 2000;279:H1087.

15. Kifor I, Dzau VJ. Endothelial renin-angiotensin pathway: Evi-dence for intracellular synthesis and secretion of angiotensins.Circ Res 1987;60:422.

16. Kambayashi B, Takahashi K, Bardhan S, Unagami T. Mo-lecular structure and function of angiotensin type 2 receptor.Kidney Intl 1994;46:1502.

17. Newton CR, Curran B, Victorino GP. Angiotensin II type 1receptor activation increases microvascular hydraulic perme-ability. Surgery 2004;136:1054.

18. Barret PQ, Bollag WB, Isales CM, McCarthy RT, Rasmussen H.Role of calcium in angiotensin II-mediated aldosterone secre-tion. Endocr Rev 1989;10:496.

19. Marrero MB, Schieffer B, Paxton WG, et al. Direct stimulationof Jak/STAT pathway by the angiotensin II AT1 Receptor.Nature 1995;375:247.

20. Michel CC, Curry FE. Microvascular permeability. Physiol Rev1999;79:703.

21. Buisson B, Laflamme L, Bottari SP, de Gasparo M, Gallo-PayetN, Payet MD. A G protein is involved in the angiotensin AT2receptor inhibition of the T-type calcium current in non-differentiated NG108-15 cells. J Biol Chem 1995;270:1670.

22. Grace PA, Mathie RT. Ischemia-reperfusion injury. London:Blackwell Science, 1999.

23. Newton CR, Curran B, Victorino GP. Angiotensin II type 1receptor activation increases microvascular permeability via acalcium dependent process. J Surg Res 2005;123:33.

24. Dennis EA, Rhee SG, Billah MM, Hannun YA. Role of phospho-lipases in generating lipid second messengers in signal trans-duction. FASEB J 1991;5:2068.

25. Garcia-Caballero A, Olivares-Reyes JA, Catt KJ, Garcia-SainzJA. Angiotensin AT1 receptor phosphorylation and desensitiza-tion in a hepatic cell line. Roles of protein kinase C and phos-phoinositide 3-kinase. Mol Pharmacol 2001;59:576.

26. He P, Zeng M, Curry FE. Dominant role of cAMP in regulationof microvessel permeability. Am J Physiol Heart Circ Physiol2000;278:H1124.

27. Neary P, Redmond HP. Ischaemia-reperfusion injury and thesystemic inflammatory response syndrome. In: Grace PA, MathieRT, eds. Ischemia-reperfusion injury. Blackwell Science: Lon-

don 1999:123.